Fusion Driven Dynamic Space-Time Clustering for Sensor Networks

Abstract

Sensors require physical interaction with the sensed phenomena and are subject to a number of noise factors. Moreover, sensor data is highly correlated across a subset of sensors in the vicinity of a stimulus. To get reliable performance from individually less reliable sensors, time-critical collaborative inference in the vicinity of a stimulus is necessary to circumvent limitations of sensing, communications, power, and equipment faults. Dynamic space-time clustering (DSTC) is the ability of a sensor network to support such collaborative inferencing in the presence of physical stimuli. In this paper we present the DSTC algorithm for tracking events and targets by deploying a sensor field in the surveillance region. The computationally efficient DSTC algorithm leverages its performance by facilitating collaboration between sensors by way of sensor data fusion. To conserve network bandwidth required for sensor data fusion, we use a probabilistic finite state-machine model based on the symbolic dynamics theory to extract useful information from time series data that represents the raw sensor data. Such a model is capable of extracting maximum useful information in the form a probabilistic finite state machine. This paper describes protocols required to carry out DSTC and to adaptively reconfigure the network, in-situ, to capture the statistical characteristics of emerging change in the information dynamics. Building on this framework, we present an urban area application that requires adaptive sensor networks to dynamically cluster sensing, processing and communications resources in space-time neighborhoods of emergent hotspots for progressively fine grained sampling and prediction, and collaborate with other dynamic clusters for event tracking.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jan 01, 2008
Accession Number
ADA518138

Entities

People

  • Bharat B. Madan
  • John Kock
  • Shashi Phoha

Organizations

  • Pennsylvania State University

Tags

Communities of Interest

  • Energy and Power Technologies
  • Sensors

DTIC Thesaurus Topics

  • Algorithms
  • Clustering
  • Control Systems
  • Data Compression
  • Data Fusion
  • Detectors
  • Formal Languages
  • Hidden Markov Models
  • Information Processing
  • Information Science
  • Network Protocols
  • Network Topology
  • Networks
  • Probability Distributions
  • Sensor Networks
  • Urban Areas
  • Wireless Sensor Networks

Fields of Study

  • Computer science

Readers

  • Agent-Based Social Robotics and Mobile-Assisted Learning in Virtual Environments.
  • Artificial Intelligence
  • Sensor Fusion and Tracking Systems.

Technology Areas

  • AI & ML
  • Space
  • Space - Space Objects